This content is not included in
your SAE MOBILUS subscription, or you are not logged in.
Massively Parallel Multifrontal Methods for Finite Element Analysis on MIMD Computer Systems
Annotation ability available
Sector:
Language:
English
Abstract
The development of highly parallel direct solvers for large, sparse linear systems of equations (e.g. for finite element or finite difference models) is lagging behind progress in parallel direct solvers for dense matrices and iterative methods for sparse matrices. We describe a massively parallel (MP) multifrontal solver for the direct solution of large sparse linear systems, such as those routinely encountered in finite element structural analysis, in an effort to address concerns about the viability of scalable, MP direct methods for sparse systems and enhance the software base for MP applications. Performance results are presented and future directions are outlined for research and development efforts in parallel multifrontal and related solvers. In particular, parallel efficiencies of 25% on 1024 nCUBE 2 nodes and 36% on 64 Intel iPSC860 nodes have been demonstrated, and parallel efficiencies of 60-85% are expected when a severe load imbalance is overcome by static mapping and dynamic load balance techniques previously developed for other parallel solvers and application codes.
Authors
Citation
Benner, R., "Massively Parallel Multifrontal Methods for Finite Element Analysis on MIMD Computer Systems," SAE Technical Paper 921083, 1992, https://doi.org/10.4271/921083.Also In
References
- Benner R. E. Montry G. R. Weigand G. G. Concurrent Multifrontal Methods: Shared Memory, Cache, and Frontwidth Issues Internat. J. Supercomputer Appls 1 3 1987 26 44
- Benner R. E. Parallel Graphics Algorithms on a 1024-Processor Hypercube Proc. 4th Conf. Hypercube Concurrent Comput. Appls. 1989 133 140
- Dragon K. M. Gustafson J. L. A Low-Cost Hypercube Load Balance Algorithm Proc. 4th Conf. Hypercube Concurrent Comput. Appls. 1989 583 590
- Duff I. S. Reid J. K. The Multifrontal Solution of Indefinite Sparse Symmetric Linear Equations ACM Trans. Math. Soft. 9 1983 302 325
- Farhat C. Wilson E. L. A New Finite Element Concurrent Computer Program Architecture Internat. J. Numer. Meth. Engr. 1987 1771 1792
- Fox G. C. Johnson M. A. Lyzenga M. A. Otto S. W. Salmon J. K. Walker D. W. Solving Problems on Concurrent Processors, Volume 1, General Techniques and Regular Problems Prentice Hall New York 1988
- Gustafson J. L. Benner R. E. Sears M. P. Sullivan T. D. A Radar Simulation Program for a 1024 Processor Hypercube Proc. Supercomputing '89 ACM Press New York 1989 96 105
- Gustafson J. L. Montry G. R. Benner R. E. Development of Parallel Methods for a 1024 Processor Hypercube SIAM J. Sci. Stat. Comput. 9 1988 609 638
- Heath M. T. Ng E. Peyton B. W. Parallel Algorithms for Sparse Linear Systems SIAM Rev. 33 1991 420 460
- Hendrickson B. Parallel QR Factorization on a Hypercube Using the Torus Wrap Mapping Sandia Report SAND91-0874 Sandia National Laboratories Albuquerque, NM January 1992
- Hood P. Frontal Solution Program for Unsymmetric Matrices Internat. J. Numer. Meth. Engr. 2 1970 5 32
- Irons B. M. A Frontal Solution Program for Finite Element Analysis Internat. J. Numer. Meth. Engr. 10 1976 379 399 11 1977 1055
- Liu J. W. H. The Multifrontal Method for Sparse Matrix Solution: Theory and Practice SIAM Rev. 34 1992 82 109
- Lucas R. F. Solving Planar Systems of Equations on Distributed Memory Multiprocessors Stanford University Dec. 1987
- Morison R. Otto S. W. The Scattered Decomposition for Finite Element Problems J. Sci. Comput. 2 1986
- Sears M. P. Linear Algebra for Dense Matrices on a Hypercube Proc. Fifth Distrib. Mem. Comput. Conf. 1990 317 320
- Simon H. Partitioning of Unstructured Problems for Parallel Processing Comput. Sys. Engr. 2 1991 135 148
- Vaughan C. T. Implementation of JAC3D on the NCUBE/ten Proc. Fifth Distrib. Mem. Comput. Conf. 1990 538 544
- Vaughan C. T. Structural Analysis on Massively Parallel Computers Comput. Sys. Engr. 2 1991 261 268
- Venugopal S. Naik V. K. Effects of Partitioning and Scheduling Sparse Matrix Factorization on Communication and Load Balance ICASE Report 91-80 NASA Langley Research Center Hampton, Va. October 1991